US5147817A - Method for forming programmable resistive element - Google Patents

Method for forming programmable resistive element Download PDF

Info

Publication number
US5147817A
US5147817A US07/614,889 US61488990A US5147817A US 5147817 A US5147817 A US 5147817A US 61488990 A US61488990 A US 61488990A US 5147817 A US5147817 A US 5147817A
Authority
US
United States
Prior art keywords
forming
layer
channel
disposed outwardly
barrier
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
US07/614,889
Inventor
Gary A. Frazier
Robert T. Bate
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Texas Instruments Inc
Original Assignee
Texas Instruments Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Texas Instruments Inc filed Critical Texas Instruments Inc
Priority to US07/614,889 priority Critical patent/US5147817A/en
Assigned to TEXAS INSTRUMENTS INCORPORATED, A CORP. OF DELAWARE reassignment TEXAS INSTRUMENTS INCORPORATED, A CORP. OF DELAWARE ASSIGNMENT OF ASSIGNORS INTEREST. Assignors: BATE, ROBERT T., FRAZIER, GARY A.
Priority to JP30064691A priority patent/JP3222166B2/en
Application granted granted Critical
Publication of US5147817A publication Critical patent/US5147817A/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/06Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
    • G06N3/063Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06GANALOGUE COMPUTERS
    • G06G7/00Devices in which the computing operation is performed by varying electric or magnetic quantities
    • G06G7/48Analogue computers for specific processes, systems or devices, e.g. simulators
    • G06G7/62Analogue computers for specific processes, systems or devices, e.g. simulators for electric systems or apparatus
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S438/00Semiconductor device manufacturing: process
    • Y10S438/936Graded energy gap

Definitions

  • the present invention relates in general to resistive elements and in particular to a programmable resistive element.
  • Neural networks provide a means for solving random problems, such as in real-world sensing systems which must "learn” from the surrounding environment so that solutions can improve with experience. Neural networks solve these sensing problems by expressing the sensor outputs as multi-dimensional vectors and then "learn" the vectors by constructing a matrix by correlation methods.
  • Neural networks use arrays formed by rows and columns of weighing elements, represented by resistors, to create the matrix vectors of voltages input from corresponding sensors using Ohm's law. Operational amplifiers sum the currents resulting from the drop of an input voltage across the resistors in each of the rows. The current output from each row represents the vector product for one component of a corresponding output vector.
  • resistors By storing the data in terms of resistor conductance values, an environment can be "learned” and later retrieved by associative recall. Thus, the resistors must first change with the system experience to "learn" an environment, but then remain fixed to recall the stored environment. Further, the neural network system can optimize the learned patterns of various environments by varying the resistive elements in the network.
  • a programmable resistive element which includes a storage gate comprising of a layer of intrinsic gallium arsenide.
  • the storage gate is separated from a channel comprising a layer of gallium arsenide by a programming barrier.
  • An insulator layer is outwardly disposed from the storage gate.
  • a control gate is disposed outwardly from the insulator. First and second spaced apart contacts, contact the channel.
  • a voltage is applied to the control gate to induce an enhancement region in the channel.
  • Application of a voltage between the first and second spaced apart contacts established a current flow through the enhancement region in the channel.
  • a number of electrons are injected onto the storage gate thereby programming the programmable resistive element.
  • the programming barrier is trapezoidal such that electrons can be injected onto the trapezoidal barrier when a suitable voltage is applied to the gate. While electrons are prevented from escaping from the storage gate when no voltage is applied to a control gate.
  • the present invention provides a programmable resistive element which may be used as either a non-volatile memory or as the resistive elements in systems such neural networks and the resistive ladders used in analog to digital converters.
  • the present invention has the advantages of high speed, the ability to operate over a large temperature range and good immunity to radiation. Further, the use of a trapezoidal barrier allows the retention of carriers while at the same time not providing substantial difficulty in programming the device.
  • FIG. 1 is a schematic elevational sectional view of a programmable resistive device according to the invention
  • FIG. 2 is a band profile drawing of the programmable resistive element of FIG. 1;
  • FIG. 3 is a band profile drawing of a first alternate embodiment of the programmable resistive element of FIG. 1;
  • FIG. 4 is a band profile drawing of a second alternate embodiment of the programmable resistive element of FIG. 1;
  • FIG. 5 is a schematic elevational sectional view of a preferred embodiment of a programmable resistive element having improved programming capabilities according to the invention
  • FIG. 6 is a top view of the programmable resistive element shown in FIG. 5;
  • FIG. 7 is a band profile drawing of the programmable resistive element of FIG. 5.
  • a programmable gallium arsenide metal semiconductor field effect transistor is shown generally at 10.
  • the transistor has a vertical structure and is formed on a semi-insulating gallium arsenide substrate 12.
  • the vertical structure can be formed, for example, by successive layer growth by well-known techniques such as molecular beam epitaxy.
  • a buffer 14 is grown across the face of substrate 12 and may be, for example, formed by a superlattice or a layer of wide bandgap material such as AlGaAs.
  • a layer 16 of n-GaAs is next formed across buffer 14.
  • Layer 16 is grown to a thickness of 3500 to 5500 angstroms and has a dopant concentration in the range of 1 ⁇ 10 17 cm -3 .
  • Layer 14 functions as the channel of field effect transistor 10.
  • a layer 18 of AlGaAs is formed across layer 16.
  • Layer 18 has a thickness of 700 to 1300 angstroms and functions as the programming barrier in transistor 10.
  • a layer 20 of intrinsic gallium arsenide (i-GaAs) is next formed across the surface of AlGaAs layer 18 to be of a thickness of 700 to 1300 angstroms.
  • Intrinsic gallium arsenide layer 20 functions as the storage gate of transistor 10.
  • a second insulating layer 22 of AlGaAs is then formed to a thickness of 700 to 1300 angstroms across the upper surface of floating gate 20.
  • Gate 24 is formed of a layer of metal across AlGaAs layer 22 and controls both the conductivity of channel 16 and the injection of carriers onto floating gate 20.
  • gate 24 functions as a Schottky barrier.
  • a metal drain contact 26 and a metal source contact 28 are formed to provide ohmic contacts with channel area 16. Drain contact 26 and contact 28 operate to induce current flow in channel 16 as with a conventional MESFET.
  • the sides of intrinsic gallium arsenide 20 are separated from source contact 28 and drain contact 26 by insulating regions (not shown), such as regions of aluminum nitride to prevent carrier transport.
  • transistor 10 can now be described referring to FIG. 1, and to FIG. 2, which depicts the band profile for transistor 10.
  • a gate voltage V g of 5 volts is applied to gate 24.
  • the application of gate voltage V g creates an enhancement region in channel 16.
  • Current flow is established through the enhancement region in channel 16 by applying a voltage V ds between drain 26 and source 28.
  • source 28 is held at zero volts or ground while a voltage of 5 is applied to drain region 26.
  • a number of the electrons moving between the source and the drain will have sufficient energy to inject hot electrons across AlGaAs barrier layer 18. These "hot" electrons come to reside on i-GaAs storage gate 20.
  • the accumulation of electrons on storage gate 20 reduces the electric field created in the channel 16 when a reading voltage of 2 volts is applied to gate 24.
  • the reduced electric field in channel 16 narrows the enhancement region thereby increasing the impedance between drain 26 and source 28 due to velocity saturation in the pinch-off region near drain 26.
  • programmable MESFET 10 can be employed as a nonvolatile memory similar to conventional EEPROM cells.
  • barrier layer 18 is formed by a graded or modulated composition of AlGaAs. As barrier layer 18 is formed adjacent channel 16, the percentage of aluminum atoms replacing gallium atoms in the lattice is gradually increased.
  • barrier 18 may comprise Al 0 .20 Ga 0 .80 As adjacent channel 16 and Al 0 .40 Ga 0 .60 As adjacent floating gate 20.
  • Such a compositionally graded structure may be formed, for example, by molecular beam epitaxy. While the preferred embodiment utilized aluminum-gallium-arsenide to form insulating layers, other materials are available for the same use. Examples include aluminum-gallium-arsenide, indium-arsenide gallium-nitride, an aluminum-arsenide.
  • barrier layer 18 results in a trapezoidal barrier which separates channel 16 from floating gate 20.
  • the shape of the barrier is such that injecting electrons onto storage gate 20 requires mounting a relatively low barrier, since as the programming voltage is applied to gate 24, the trapezoid "flattens out", and electrons are transferred into storage gate 20.
  • the voltage is subsequently removed from gate 24, it is more difficult for electrons to escape from storage gate 20, since the barrier height in the reverse direction is much higher than that of the forward direction. This allows electrons to be retained in storage gate 20.
  • a large reverse bias is created between gate 24 and drain 26, such that electrons can be drawn in a reverse direction over the trapezoidal barrier created by barrier layer 18.
  • barrier layer 18 which presents a trapezoidal barrier between channel 16 and floating gate 20 is not limited to composition modulation as depicted in FIG. 3.
  • An alternate approach, using a chirped superlattice is depicted in the band profile of FIG. 4.
  • barrier layer 18 is replaced with the superlattice in which the bottom of the first conduction band mini-zone is the position of the effective barrier height.
  • the spacial frequency of the super-lattice is increased ("chirped") such that the edge of the mini-zone rises in real space.
  • Such a superlattice structure can be formed, for example, using a superlattice emulator.
  • FIG. 5 a second embodiment of transistor 10 is shown in which floating gate 20 may be charged with either electrons or holes.
  • a top plan view of the transistor shown in FIG. 5 is given in FIG. 6.
  • a programming channel 30 has been added to transistor 10.
  • Programming channel 30 is separated from channel 16 by a layer of AlGaAs 32 of a thickness of 700 to 1300 angstroms.
  • An n-type contact 34 and a p-type contact 36 are made to opposite sides of the programming channel 30 as shown in FIG. 6.
  • Contacts 34 and 36 may be formed for example by doped gallium arsenide.
  • Barrier layer 16, now disposed between programming channel 30 and floating gate 20, is preferably graded to form a trapezoidal barrier using one of the approaches described above.
  • the band profile for the structure shown in FIGS. 5 and 6 is given in FIG. 7.
  • programming channel 30 and contacts 34 and 36 allow floating gate 20 to be selectively charged with either electrons or holes. This process of selective injection is possible because the lateral p-i-n structure created by the n-type contact 34, the p-type contact 36 and the intrinsic layer comprising programming channel 30.
  • the electrons or holes are injected from programming channel 30 to storage gate 20 when carrier flow is established between n-type contact 34 and p-type contact 36. Electrons may be drawn into channel 30 from n-type contact 34 while holes may be drawn into the channel from p-type contact.
  • Channel 16 acts as a ground plane in close proximity to programming channel 30 allowing the control of the electrostatic potential of programming channel 30.
  • the p-i voltage and the i-n voltage can be controlled separately.
  • the magnitude of the charge injected onto storage gate 20 can be controlled by varying the terminal current applied to n-type contact 34 and p-type contact 36, thereby setting the impedance of channel 16.

Abstract

A programmable resistive element is provided which includes a channel 16 comprising a layer of gallium arsenide. A programming barrier 18 is disposed outwardly from channel 16. A storage gate 20 comprising a layer of intrinsic gallium arsenide is disposed outwardly from programming barrier 18. An insulator 22, comprising a layer of aluminum-gallium-arsenide, is disposed outwardly from storage gate 20. A control gate 24 is disposed outwardly from insulator 22. First and second spaced apart contacts 26 and 28 contact channel 16.

Description

TECHNICAL FIELD OF THE INVENTION
The present invention relates in general to resistive elements and in particular to a programmable resistive element.
BACKGROUND OF THE INVENTION
Programmable resistors can be utilized in a number of analog signal processing applications such as resistive ladders in analog to digital converters and in resistor arrays in neural networks. Neural networks provide a means for solving random problems, such as in real-world sensing systems which must "learn" from the surrounding environment so that solutions can improve with experience. Neural networks solve these sensing problems by expressing the sensor outputs as multi-dimensional vectors and then "learn" the vectors by constructing a matrix by correlation methods.
Neural networks use arrays formed by rows and columns of weighing elements, represented by resistors, to create the matrix vectors of voltages input from corresponding sensors using Ohm's law. Operational amplifiers sum the currents resulting from the drop of an input voltage across the resistors in each of the rows. The current output from each row represents the vector product for one component of a corresponding output vector. By storing the data in terms of resistor conductance values, an environment can be "learned" and later retrieved by associative recall. Thus, the resistors must first change with the system experience to "learn" an environment, but then remain fixed to recall the stored environment. Further, the neural network system can optimize the learned patterns of various environments by varying the resistive elements in the network.
Various means have been devised for providing variable weighing elements in neural networks. Each of these means has been found to have significant disadvantages. For example, circuitry using up-down counters and decoded switches along with fixed resistors may be used to generate the appropriate resistive weights, however, such an approach would be limited to only a small number of weighing elements, thereby limiting the complexity and utility of corresponding application. Electrically-erasable, electrically-programmable read-only memories (EEPROM) employing metal oxide semiconductor field effect transistors (MOSFETs) based on well known silicon structures provide another option. Silicon based devices, however, have significant limitations as to switching speed erasability lifetime, radiation tolerance and high temperature operation. Finally, dynamic random-access memories (DRAMs) have been considered, but DRAMs need refreshing after a read operation which greatly increases the number of overhead operations required in the overall scheme of the application.
Thus, the need has arisen for an improved programmable resistor having the capability to operate in demanding applications, such as neural networks.
SUMMARY OF THE INVENTION
According to the invention, a programmable resistive element is provided which includes a storage gate comprising of a layer of intrinsic gallium arsenide. The storage gate is separated from a channel comprising a layer of gallium arsenide by a programming barrier. An insulator layer is outwardly disposed from the storage gate. A control gate is disposed outwardly from the insulator. First and second spaced apart contacts, contact the channel.
According to other aspects of the invention, a voltage is applied to the control gate to induce an enhancement region in the channel. Application of a voltage between the first and second spaced apart contacts established a current flow through the enhancement region in the channel. A number of electrons are injected onto the storage gate thereby programming the programmable resistive element.
According to further aspects of the invention, the programming barrier is trapezoidal such that electrons can be injected onto the trapezoidal barrier when a suitable voltage is applied to the gate. While electrons are prevented from escaping from the storage gate when no voltage is applied to a control gate.
The present invention provides a programmable resistive element which may be used as either a non-volatile memory or as the resistive elements in systems such neural networks and the resistive ladders used in analog to digital converters. The present invention has the advantages of high speed, the ability to operate over a large temperature range and good immunity to radiation. Further, the use of a trapezoidal barrier allows the retention of carriers while at the same time not providing substantial difficulty in programming the device.
BRIEF DESCRIPTION OF THE DRAWINGS
Other aspects of the invention and their advantages will be discerned when one refers to the following detailed description as taken in conjunction with the drawings, in which like numbers identify like parts and in which:
FIG. 1 is a schematic elevational sectional view of a programmable resistive device according to the invention;
FIG. 2 is a band profile drawing of the programmable resistive element of FIG. 1;
FIG. 3 is a band profile drawing of a first alternate embodiment of the programmable resistive element of FIG. 1;
FIG. 4 is a band profile drawing of a second alternate embodiment of the programmable resistive element of FIG. 1;
FIG. 5 is a schematic elevational sectional view of a preferred embodiment of a programmable resistive element having improved programming capabilities according to the invention;
FIG. 6 is a top view of the programmable resistive element shown in FIG. 5; and
FIG. 7 is a band profile drawing of the programmable resistive element of FIG. 5.
DETAILED DESCRIPTION OF THE INVENTION
Referring first to FIG. 1, a programmable gallium arsenide metal semiconductor field effect transistor (MESFET) is shown generally at 10. The transistor has a vertical structure and is formed on a semi-insulating gallium arsenide substrate 12. The vertical structure can be formed, for example, by successive layer growth by well-known techniques such as molecular beam epitaxy. A buffer 14 is grown across the face of substrate 12 and may be, for example, formed by a superlattice or a layer of wide bandgap material such as AlGaAs. A layer 16 of n-GaAs is next formed across buffer 14. Layer 16 is grown to a thickness of 3500 to 5500 angstroms and has a dopant concentration in the range of 1×1017 cm-3. Layer 14 functions as the channel of field effect transistor 10.
Next, a layer 18 of AlGaAs is formed across layer 16. Layer 18 has a thickness of 700 to 1300 angstroms and functions as the programming barrier in transistor 10. A layer 20 of intrinsic gallium arsenide (i-GaAs) is next formed across the surface of AlGaAs layer 18 to be of a thickness of 700 to 1300 angstroms. Intrinsic gallium arsenide layer 20 functions as the storage gate of transistor 10. A second insulating layer 22 of AlGaAs is then formed to a thickness of 700 to 1300 angstroms across the upper surface of floating gate 20.
Gate 24 is formed of a layer of metal across AlGaAs layer 22 and controls both the conductivity of channel 16 and the injection of carriers onto floating gate 20. Preferably, gate 24 functions as a Schottky barrier. A metal drain contact 26 and a metal source contact 28 are formed to provide ohmic contacts with channel area 16. Drain contact 26 and contact 28 operate to induce current flow in channel 16 as with a conventional MESFET. In the preferred embodiment, the sides of intrinsic gallium arsenide 20 are separated from source contact 28 and drain contact 26 by insulating regions (not shown), such as regions of aluminum nitride to prevent carrier transport.
The operation of transistor 10 can now be described referring to FIG. 1, and to FIG. 2, which depicts the band profile for transistor 10. In operation, a gate voltage Vg of 5 volts is applied to gate 24. The application of gate voltage Vg creates an enhancement region in channel 16. Current flow is established through the enhancement region in channel 16 by applying a voltage Vds between drain 26 and source 28. Preferably, source 28 is held at zero volts or ground while a voltage of 5 is applied to drain region 26. A number of the electrons moving between the source and the drain will have sufficient energy to inject hot electrons across AlGaAs barrier layer 18. These "hot" electrons come to reside on i-GaAs storage gate 20.
The accumulation of electrons on storage gate 20 reduces the electric field created in the channel 16 when a reading voltage of 2 volts is applied to gate 24. The reduced electric field in channel 16 narrows the enhancement region thereby increasing the impedance between drain 26 and source 28 due to velocity saturation in the pinch-off region near drain 26.
By storing electrons in storage gate 20, and consequently changing the impedance of channel 16, the present invention can be employed in at least two different ways. First, programmable MESFET 10 can be employed as a nonvolatile memory similar to conventional EEPROM cells. Second, by controlling the amount of electrons injected onto storage gate 20, the resistance of channel 16 can be effectively controlled, allowing programmable MESFET 10 to be used as a programmable resistive element.
Referring next to FIG. 3, a band profile for an alternative structure of transistor 10 is depicted. In this embodiment, barrier layer 18 is formed by a graded or modulated composition of AlGaAs. As barrier layer 18 is formed adjacent channel 16, the percentage of aluminum atoms replacing gallium atoms in the lattice is gradually increased. For example, barrier 18 may comprise Al0.20 Ga0.80 As adjacent channel 16 and Al0.40 Ga0.60 As adjacent floating gate 20. Such a compositionally graded structure may be formed, for example, by molecular beam epitaxy. While the preferred embodiment utilized aluminum-gallium-arsenide to form insulating layers, other materials are available for the same use. Examples include aluminum-gallium-arsenide, indium-arsenide gallium-nitride, an aluminum-arsenide.
Grading the composition of barrier layer 18 results in a trapezoidal barrier which separates channel 16 from floating gate 20. The shape of the barrier is such that injecting electrons onto storage gate 20 requires mounting a relatively low barrier, since as the programming voltage is applied to gate 24, the trapezoid "flattens out", and electrons are transferred into storage gate 20. When the voltage is subsequently removed from gate 24, it is more difficult for electrons to escape from storage gate 20, since the barrier height in the reverse direction is much higher than that of the forward direction. This allows electrons to be retained in storage gate 20. To remove electrons from storage gate 20, a large reverse bias is created between gate 24 and drain 26, such that electrons can be drawn in a reverse direction over the trapezoidal barrier created by barrier layer 18.
Formation of a barrier layer 18 which presents a trapezoidal barrier between channel 16 and floating gate 20 is not limited to composition modulation as depicted in FIG. 3. An alternate approach, using a chirped superlattice is depicted in the band profile of FIG. 4. In this embodiment, barrier layer 18 is replaced with the superlattice in which the bottom of the first conduction band mini-zone is the position of the effective barrier height. The spacial frequency of the super-lattice is increased ("chirped") such that the edge of the mini-zone rises in real space. Such a superlattice structure can be formed, for example, using a superlattice emulator.
Referring next to FIG. 5, a second embodiment of transistor 10 is shown in which floating gate 20 may be charged with either electrons or holes. A top plan view of the transistor shown in FIG. 5 is given in FIG. 6. In this embodiment, a programming channel 30 has been added to transistor 10. Programming channel 30 is separated from channel 16 by a layer of AlGaAs 32 of a thickness of 700 to 1300 angstroms. An n-type contact 34 and a p-type contact 36 are made to opposite sides of the programming channel 30 as shown in FIG. 6. Contacts 34 and 36 may be formed for example by doped gallium arsenide. Barrier layer 16, now disposed between programming channel 30 and floating gate 20, is preferably graded to form a trapezoidal barrier using one of the approaches described above. The band profile for the structure shown in FIGS. 5 and 6 is given in FIG. 7.
The addition of programming channel 30 and contacts 34 and 36 allow floating gate 20 to be selectively charged with either electrons or holes. This process of selective injection is possible because the lateral p-i-n structure created by the n-type contact 34, the p-type contact 36 and the intrinsic layer comprising programming channel 30. The electrons or holes are injected from programming channel 30 to storage gate 20 when carrier flow is established between n-type contact 34 and p-type contact 36. Electrons may be drawn into channel 30 from n-type contact 34 while holes may be drawn into the channel from p-type contact. Channel 16 acts as a ground plane in close proximity to programming channel 30 allowing the control of the electrostatic potential of programming channel 30. Thus, the p-i voltage and the i-n voltage can be controlled separately. The magnitude of the charge injected onto storage gate 20 can be controlled by varying the terminal current applied to n-type contact 34 and p-type contact 36, thereby setting the impedance of channel 16.
Although a preferred embodiment of the present invention has been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (2)

What is claimed is:
1. A method for forming a programmable resistive element, comprising the steps of:
forming a channel region comprising a layer of gallium arsenide;
forming a trapezoidal programming barrier disposed outwardly from said channel region;
forming a storage gate disposed outwardly from said programming barrier comprising a layer of intrinsic gallium arsenide;
forming an insulator disposed outwardly from said storage gate;
forming a control gate disposed outwardly from said insulator; and
forming first and second spaced apart contacts to contact said channel, wherein said step of forming a programming barrier comprised the step of forming a layer of composition modulated aluminum-gallium-arsenide.
2. A method for forming a programmable resistive element, comprising the steps of:
forming a channel region comprising a layer of gallium arsenide;
forming a trapezoidal programming barrier disposed outwardly from said channel region;
forming a storage gate disposed outwardly from said programming barrier comprising a layer of intrinsic gallium arsenide;
forming an insulator disposed outwardly from said storage gate;
forming a control gate disposed outwardly from said insulator; and
forming first and second spaced apart contacts to contact said channel, wherein said step of forming a trapezoidal barrier comprises the step of forming a chirped superlattice.
US07/614,889 1990-11-16 1990-11-16 Method for forming programmable resistive element Expired - Lifetime US5147817A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US07/614,889 US5147817A (en) 1990-11-16 1990-11-16 Method for forming programmable resistive element
JP30064691A JP3222166B2 (en) 1990-11-16 1991-11-15 Programmable resistance element

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US07/614,889 US5147817A (en) 1990-11-16 1990-11-16 Method for forming programmable resistive element

Publications (1)

Publication Number Publication Date
US5147817A true US5147817A (en) 1992-09-15

Family

ID=24463128

Family Applications (1)

Application Number Title Priority Date Filing Date
US07/614,889 Expired - Lifetime US5147817A (en) 1990-11-16 1990-11-16 Method for forming programmable resistive element

Country Status (2)

Country Link
US (1) US5147817A (en)
JP (1) JP3222166B2 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5296395A (en) * 1991-05-17 1994-03-22 Apa Optics, Inc. Method of making a high electron mobility transistor
US6023079A (en) * 1995-07-07 2000-02-08 Nec Corporation Compound semiconductor memory with floating gate
US20040195562A1 (en) * 2002-11-25 2004-10-07 Apa Optics, Inc. Super lattice modification of overlying transistor
DE102011006782A1 (en) * 2011-04-05 2012-10-11 Technische Universität Berlin memory cell

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4558337A (en) * 1984-05-30 1985-12-10 Texas Instruments Inc. Multiple high electron mobility transistor structures without inverted heterojunctions
JPS6169476A (en) * 1984-09-14 1986-04-10 Canon Inc Image processor
EP0202755A1 (en) * 1985-04-19 1986-11-26 International Business Machines Corporation Double heterojunction field effect transistor data storage device
EP0305253A1 (en) * 1987-08-05 1989-03-01 Thomson-Csf Semiconducteurs Specifiques Double heterojunction microwave transistor
US4905063A (en) * 1988-06-21 1990-02-27 American Telephone And Telegraph Company, At&T Bell Laboratories Floating gate memories
US4945393A (en) * 1988-06-21 1990-07-31 At&T Bell Laboratories Floating gate memory circuit and apparatus
US5023674A (en) * 1985-08-20 1991-06-11 Fujitsu Limited Field effect transistor
US5038187A (en) * 1989-12-01 1991-08-06 Hewlett-Packard Company Pseudomorphic MODFET structure having improved linear power performance at microwave frequencies
US5055890A (en) * 1990-01-25 1991-10-08 The United States Of America As Represented By The United States Department Of Energy Nonvolatile semiconductor memory having three dimension charge confinement

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4558337A (en) * 1984-05-30 1985-12-10 Texas Instruments Inc. Multiple high electron mobility transistor structures without inverted heterojunctions
JPS6169476A (en) * 1984-09-14 1986-04-10 Canon Inc Image processor
EP0202755A1 (en) * 1985-04-19 1986-11-26 International Business Machines Corporation Double heterojunction field effect transistor data storage device
US5023674A (en) * 1985-08-20 1991-06-11 Fujitsu Limited Field effect transistor
EP0305253A1 (en) * 1987-08-05 1989-03-01 Thomson-Csf Semiconducteurs Specifiques Double heterojunction microwave transistor
US4905063A (en) * 1988-06-21 1990-02-27 American Telephone And Telegraph Company, At&T Bell Laboratories Floating gate memories
US4945393A (en) * 1988-06-21 1990-07-31 At&T Bell Laboratories Floating gate memory circuit and apparatus
US5038187A (en) * 1989-12-01 1991-08-06 Hewlett-Packard Company Pseudomorphic MODFET structure having improved linear power performance at microwave frequencies
US5055890A (en) * 1990-01-25 1991-10-08 The United States Of America As Represented By The United States Department Of Energy Nonvolatile semiconductor memory having three dimension charge confinement

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Brodsky et al., "GaAs Electrically Alterable Read-Only Memory", IBM Tech. Disc. Bull., vol. 25, No. 3A, Aug. 1982, p. 1250.
Brodsky et al., GaAs Electrically Alterable Read Only Memory , IBM Tech. Disc. Bull., vol. 25, No. 3A, Aug. 1982, p. 1250. *
Kostalsky et al., "Novel High-Speed Transistor Based on Charge Emission from a Quantum Well", Appl. Phys. Lett., 52(11), Mar. 14, 1988, pp. 904-906.
Kostalsky et al., Novel High Speed Transistor Based on Charge Emission from a Quantum Well , Appl. Phys. Lett., 52(11), Mar. 14, 1988, pp. 904 906. *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5296395A (en) * 1991-05-17 1994-03-22 Apa Optics, Inc. Method of making a high electron mobility transistor
US6023079A (en) * 1995-07-07 2000-02-08 Nec Corporation Compound semiconductor memory with floating gate
US6025612A (en) * 1995-07-07 2000-02-15 Nec Corporation NAND or NOR compound semiconductor memory
US20040195562A1 (en) * 2002-11-25 2004-10-07 Apa Optics, Inc. Super lattice modification of overlying transistor
US7112830B2 (en) 2002-11-25 2006-09-26 Apa Enterprises, Inc. Super lattice modification of overlying transistor
DE102011006782A1 (en) * 2011-04-05 2012-10-11 Technische Universität Berlin memory cell
US9424925B2 (en) 2011-04-05 2016-08-23 Technische Universität Berlin Memory cell

Also Published As

Publication number Publication date
JPH05198765A (en) 1993-08-06
JP3222166B2 (en) 2001-10-22

Similar Documents

Publication Publication Date Title
US5065132A (en) Programmable resistor and an array of the same
US5923046A (en) Quantum dot memory cell
US4194935A (en) Method of making high mobility multilayered heterojunction devices employing modulated doping
EP0005059B1 (en) A semiconductor device having a layered structure and a method of making it
US5021841A (en) Semiconductor device with controlled negative differential resistance characteristic
US5783840A (en) Universal quantum dot logic cell
US4780748A (en) Field-effect transistor having a delta-doped ohmic contact
US4905063A (en) Floating gate memories
US4835581A (en) Electron gas hole gas tunneling transistor device
US4734750A (en) High electron mobility heterojunction semiconductor devices
US5023674A (en) Field effect transistor
US4675711A (en) Low temperature tunneling transistor
KR100542963B1 (en) Field effect transistor
US5932889A (en) Semiconductor device with floating quantum box
US5147817A (en) Method for forming programmable resistive element
FR2508707A1 (en) BALLISTIC TRANSISTOR WITH MULTIPLE HETEROJUNCTIONS
US5012301A (en) Three terminal semiconductor device
US5055890A (en) Nonvolatile semiconductor memory having three dimension charge confinement
WO1991013465A1 (en) Electronic devices and methods of constructing and utilizing same
JP2973876B2 (en) Compound semiconductor memory
JPS639388B2 (en)
EP0202755A1 (en) Double heterojunction field effect transistor data storage device
EP0545255A2 (en) Quantum semiconductor device employing quantum boxes for enabling compact size and high-speed operation
JPS6353705B2 (en)
JP2734435B2 (en) Tunnel transistor and storage circuit

Legal Events

Date Code Title Description
AS Assignment

Owner name: TEXAS INSTRUMENTS INCORPORATED, A CORP. OF DELAW

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNORS:FRAZIER, GARY A.;BATE, ROBERT T.;REEL/FRAME:005519/0275

Effective date: 19901116

STCF Information on status: patent grant

Free format text: PATENTED CASE

FPAY Fee payment

Year of fee payment: 4

FPAY Fee payment

Year of fee payment: 8

FPAY Fee payment

Year of fee payment: 12